Oral anabolic steroids produce striking reductions in serum concentrations of high-density lipoprotein (HDL) cholesterol. We hypothesized that this effect related to their route of administration and was unrelated to their androgenic potency. We administered oral stanozolol (6 mg/d) or supraphysiological doses of intramuscular testosterone enanthate (200 mg/wk) to 11 male weight lifters for six weeks in a crossover design. Stanozolol reduced HDL-cholesterol and the HDL2 subfraction by 33% and 71%, respectively. In contrast, testosterone decreased HDL-cholesterol concentration by only 9% and the decrease was in the HDL3 subfraction. Apolipoprotein A-I level decreased 40% during stanozolol but only 8% during testosterone treatment. The low-density lipoprotein cholesterol concentration increased 29% with stanozolol and decreased 16% with testosterone treatment. Stanozolol, moreover, increased postheparin hepatic triglyceride lipase activity by 123%, whereas the maximum change during testosterone therapy (+25%) was not significant. Weight gain was similar with both drugs, but testosterone was more effective in suppressing gonadotropic hormones. We conclude that the undesirable lipoprotein effects of 17-alpha-alkylated steroids given orally are different from those of parenteral testosterone and that the latter may be preferable in many clinical situations.